I don't know the list of languages included in the book, but anyone interested should try out:

   - J: https://www.jsoftware.com/#/ you'll learn arbitrary dimension arrays, verb composition and more
   - Lisp: I don't know the right Lisp to recommend. Macros stand out as something to learn
   - Forth: I don't know the right Forth to recommend. The stack is an interesting metaphor
   - A constraint-based language: maybe https://www.minizinc.org constraints are super interesting

Those are all practical languages -- none is esoteric just for the sake of strangeness.

There are more of course.

▲ userbinator 4 days ago | parent | next [-]

I don't think those languages would be "esoteric" in the meaning of the article (purposefully designed to be different for the sake of being different) but they are more like domain-specific languages designed to fit certain use-cases extremely well, and in doing so, depart from the traditional imperative model.

▲ jacquesm 4 days ago | parent | prev | next [-]

The right Forth is the one you write.

▲

Nevermark 4 days ago | parent [-]

In your Forth.

	▲	OhMeadhbh 3 days ago \| parent [-]
		And then you write a custom outer interpreter to create a DSL designed to write FORTH inner interpreters.

▲ olmo23 3 days ago | parent | prev | next [-]

Racket is the canonical LISP (well, Scheme) to recommend to new users.

https://racket-lang.org/

▲ tmtvl 3 days ago | parent [-]

Yeah, as much as I love Common Lisp and prefer it to Scheme (despite learning Scheme first), the fact you can read the entirety of R7RS small in a single afternoon is, in my opinion, quite important in making the language approachable to beginners.

	▲	gus_massa 3 days ago \| parent [-]
		If you wish, you can install a package to run R7RS small program in Racket. Just run in the commend line `raco pkg install r7rs` or search r7rs in the package manager inside DrRacket. It's not maintained by the core developers, but it pass all the test suit and if you ever find a bug and post an issue in github or in the discourse group, it will be fixed very soon. https://pkgs.racket-lang.org/package/r7rs https://github.com/lexi-lambda/racket-r7rs more info: https://github.com/lassik/racket-r7rs-example

▲ anthk 3 days ago | parent | prev | next [-]

Lisp is not esoteric. Emacs, Guix, AutoCad...

▲ UltraSane 3 days ago | parent | prev | next [-]

I would add an SMT solver like Z3. It can solve problems when they are translated into first order logic.

▲ 7thaccount 4 days ago | parent | prev | next [-]

Good choices. Add APL, Prolog, and Rebol.

	▲	lordhumphrey 3 days ago \| parent [-]
		Hmmm. GP listed three languages which are not esoteric, and you've patted them on the back and shared three more which are not esoteric. Oops. Esoteric language =/= any non-mainstream language. Here's the definition from the esoteric programming language wiki, which is a lovely resource for anyone interested https://esolangs.org/wiki/Esoteric_programming_language: > An esoteric programming language is a computer programming language designed to experiment with weird ideas, to be hard to program in, or as a joke, rather than for practical use.

▲ hungmung 3 days ago | parent | prev [-]

Forth would be neat on an RPN calculator.

/random